State v. Inman

Defendant challenges his convictions for first-degree sodomy, a violation of ORS 163.405, and harassment, a violation of ORS 166.065. He first argues that the trial court erred by “permitting” a police detective to comment on the credibility of another witness, the sodomy victim, despite the lack of any objection by defendant to that aspect of the detective’s testimony. As explained below, even if we agreed that the trial court plainly erred by not striking that testimony sua sponte, we would decline to exercise our discretion to correct the error. Accordingly, we need not determine whether the trial court plainly erred in the way that defendant contends. Defendant also argues that the trial court erred when it made various rulings related to defendant’s conviction by nonunanimous jury verdicts (11-1 on sodomy; 10-2 on harassment) and when it imposed the 300-month statutorily mandated sentence on his sodomy conviction. We reject each of those latter arguments without further discussion. Accordingly, we affirm.

In addressing defendant’s argument that the trial court plainly erred by failing to strike certain testimony sua sponte, we consider the evidence in the light most favorable to the state. State v. Wilson, 266 Or App 481, 482, 337 P3d 990, rev den, 356 Or 837 (2014). Later, in exploring whether we should exercise our discretion to correct any such error, we consider all of the pertinent evidence and other relevant aspects of the record. Cf. State v. Blaylock, 267 Or App 455, 456 n 1, 341 P3d 758 (2014), rev den, 357 Or 299 (2015) (explaining that “we look at all pertinent evidence” when assessing whether evidentiary error was harmless). We describe the facts in keeping with those standards.

This case involves several children, including J (the sodomy victim), E (the harassment victim, and a friend of J), R (another of J’s friends), and Q (J’s older brother). In early 2012, defendant moved into the home where R lived with her parents and siblings; he stayed there for five or six months. J and E, who then were about seven years old, and Q, who was about eleven years old, often played at R’s house during the summer. One afternoon in June or July, Q saw defendant in an upstairs bedroom in that house, bending down in front *923of J with his face near her “privates.” J’s pants were down around her ankles and she had no underwear on. Q heard defendant ask J if he could “lick her privates”; Q also heard J ask defendant to stop what he was doing. Q ran downstairs and tried to tell R’s parents what was happening, but they were occupied and did not pay attention to him. Q went home and told his mother (who is also J’s mother) what had happened, but she testified at trial that she had not understood Q and, consequently, did not immediately learn of the abuse.

On August 15, J’s mother did receive information about the abuse, which prompted her to call the sheriffs office. A deputy sheriff spoke with J and Q before the case was referred to the Brookings Police Department. The next day, J’s mother met with a detective from that department, McCourt, and told him that defendant had abused J “a couple of times.”

McCourt then contacted defendant and asked him about the allegations. Defendant told McCourt that the children had “conspired against him to make up these allegations” because “he wouldn’t play with them anymore.” Defendant also told McCourt that he had kissed J on the nose and on the cheeks, and that he might have contacted her “crotch area” when they were “playing around.” Defendant denied performing oral sex on J, telling McCourt that he “would never do anything they wouldn’t like or want.” Defendant also told McCourt that J had told him that somebody else had performed oral sex on her about a month earlier.

McCourt interviewed Q on the day after his mother first contacted the sheriffs office about the abuse. Shortly thereafter, Q was interviewed by Dotson, a forensic interviewer at the local child-advocacy center. In addition, McCourt coordinated a “pretext call” from Q to defendant. During that call, McCourt gave “information and questions to [Q] about what to say,” because the child was having difficulty talking to defendant. McCourt “kind of felt bad while [they] were doing it,” but he believed that the pretext call “was necessary to try to get as much evidence as we could.” Defendant apparently made no admissions during that call. *924Q also spoke with a prosecutor a few times in the weeks before trial.

McCourt also spoke briefly with J on the day following her mother’s report to the sheriffs office, but he “did not talk to [her] about the case at all.” That same day, Dotson conducted a forensic interview of J, who told Dotson that she was at the center because “a guy molest [her].” Dotson testified that the word “molest” is not a word that a child J’s age typically would know or use. Dotson asked J if she knew what “molest” means and J said that she had forgotten. J then told Dotson that defendant had licked her vagina with his tongue and that it happened once, at night, about a week before the interview.1 Dotson testified that about 50 percent of children J’s age use the word “vagina.” J also spoke with the prosecutor before trial.

At some point, E (one of J’s friends) disclosed that, one day while the children were playing at R’s house, defendant had grabbed her legs and looked under her dress. At trial, she testified that defendant’s actions made her feel nervous because she did not “know what [he] was going to do.” E told defendant to stop, but he did not, so she ran back to her home.

During her forensic interview, J also disclosed that, on a particular occasion when defendant and E were present, defendant looked under the two girls’ dresses. J also described another incident when defendant looked under E’s dress while he pretended to take a nap.

The state initially charged defendant with two counts of first-degree sodomy and three counts of harassment, but later dismissed some of those charges. Consequently, defendant was tried on one count of first-degree sodomy (against J) and one count of harassment (against E).

In his opening statement, defendant suggested that “the amount of contact and discussions about this” that occurred between J, Q, and their mother “before the police ever got involved” might cause the jurors to question *925whether certain standards for sex-crime investigations had been met. Q, J, Dotson, McCourt, and E then testified to the facts described above. In apparent support of an argument that the children had been interviewed too many times, defendant cross-examined Dotson about protocols for investigating child sex-abuse cases. In response, she agreed with defense counsel’s statement that, according to those guidelines, “you’re to try to minimize the number of interviews with the child.” Dotson also acknowledged that she believed that J had spoken to her mother about the abuse.

During his direct examination, McCourt was asked about conversations that he and prosecutors had with J shortly before trial. McCourt responded with the testimony that is the subject of defendant’s “vouching” argument on appeal:

“Q. * * * What was the purpose of those visits?

“A. We like to make — We like to talk to the witness again before the trial just to see, make sure they’re going to be able to — if they’re going to be able to come testify in a court, make sure they still remember, or if they have — if they decide — Let’s say that they were lying to us or they told us a fib, we like to find that out before we get to this point so we can — -we don’t have to waste your time and the court’s time. And we talked to her on the Thursday of last week.

“Q. Okay. And if there had been any inconsistencies or any problems, wouldn’t those have needed to be reported to all parties?

“A. Yes.

“Q. And you did not write any such a report or * * * note any such inconsistencies at any time?

“A. I did not.”

Defendant did not object to that testimony. The prosecutor did not linger on the topic of McCourt’s conversations with J, but instead moved to questioning McCourt about the interviews of Q and the other children.

After the state rested, defendant called J’s mother as a witness and questioned her about how she interacted with J after she learned of the abuse. In addition, defendant *926asked J’s mother about her understanding of when the abuse had occurred. Defendant also called McCourt and questioned him about the information that he had received from J’s mother on the same topic.

Defendant then testified on his own behalf. He described the relationships among the children and testified that he played with them, “swinging them around in circles” and catching them as they jumped off of things. Defendant denied ever having attempted to perform oral sex on J; he also denied having tried to look between E’s legs. On cross-examination, defendant downplayed what he had told McCourt about the children having been upset at him for not playing with them; he testified that the children had only once said that they “were going to get [defendant] for not playing with them” after he got tired, and that it had been a “very minor thing.”

Neither the state nor defendant referred to McCourt’s “vouching” testimony in the closing arguments. Rather, in its closing argument, the state first summarized the evidence and how it related to the elements of the charged crimes. The state then acknowledged some discrepancies in the children’s description of events, for example, whether the abuse occurred at night or during the day, and whether it happened only once or multiple times. The state suggested that the discrepancies could be based on defendant actually having abused J more than once, or might be based on the difficulty that the young children had in remembering and describing exactly what had happened. For his part, defendant emphasized the discrepancies in the children’s statements, J’s use of terms like “molest” and “vagina,” and the number of times that the children had spoken to adults about the abuse before trial. Defendant urged the jurors to determine, based on those considerations, that they had at least reasonable doubt about whether the abuse had occurred.

The jury convicted defendant by nonunanimous verdicts and the trial court imposed the statutorily required 300-month prison sentence on the sodomy conviction, to be followed by lifetime post-prison supervision, with a concurrent 120-day sentence on the harassment conviction. This appeal followed.

*927On appeal, defendant argues that the trial court plainly erred by “permitting” — or, more precisely, by failing to strike — McCourt’s testimony that he spoke to J the week before trial in part to make sure that she had not lied. Defendant acknowledges that he did not object to McCourt’s statements at trial, but characterizes the evidence as impermissible “vouching” testimony that the trial court had a duty to strike sua sponte. Defendant contends that the trial court’s failure to take that action constitutes plain error that we should correct by reversing the judgment and remanding for a new trial.

Our consideration of an unpreserved claim of error generally encompasses two steps. First, we determine whether the trial court plainly erred. Error is “plain” if

“(1) the error is one of law, (2) the error is ‘obvious, not reasonably in dispute,’ and (3) the error ‘appears on the face of the record,’ so that we need not ‘go outside the record to identify the error or choose between competing inferences, and the facts constituting the error are irrefutable.’ State v. Reyes-Camarena, 330 Or 431, 435, 7 P3d 522 (2000) (internal quotation marks omitted).”

State v. Corkill, 262 Or App 543, 551, 325 P3d 796, rev den, 355 Or 751 (2014).

Second, if we determine that a trial court plainly erred, we then consider whether we should exercise our discretion to correct that error.

“That discretion entails making a prudential call that takes into account an array of considerations, such as the competing interests of the parties, the nature of the case, the gravity of the error, and the ends of justice in the particular case.”

State v. Vanornum, 354 Or 614, 630, 317 P3d 889 (2013) (citations omitted).

In this case, the state concedes that McCourt’s testimony “likely was an implicit comment on [J’s] credibility.” Nonetheless, the state argues that the trial court did not plainly err in admitting that testimony for two reasons. First, the state contends, defendant opened the door to that testimony when he suggested that J had been improperly *928subjected to multiple interviews and that her claims were, therefore, suspect. Second, the state argues, defendant may have made a strategic decision not to object to the testimony, given its possible admissibility in response to defendant’s theory that J had been interviewed too many times.

We need not decide whether the trial court plainly erred in admitting McCourt’s statements because, for the reasons that follow, we would not exercise our discretion to correct any plain error that might exist on this record.

The principle that appellate courts generally will not consider unpreserved claims of error is longstanding. As the Supreme Court stated in 1944, “ [i] t is only in rare and exceptional cases that this court will notice an alleged error where no ruling has been sought from the trial judge.” Hotelling v. Walther, 174 Or 381, 385, 148 P2d 933 (1944). Somewhat more recently, the court explained in Ailes v. Portland Meadows, Inc., 312 Or 376, 382, 823 P2d 956 (1991), that an appellate court’s decision to exercise its discretion to address an unpreserved claim of error “should be made with utmost caution” because “[s]uch an action is contrary to the strong policies requiring preservation and raising of error.” Ailes provided a nonexclusive list of considerations that an appellate court should consider when deciding whether to exercise that discretion:

“the competing interests of the parties; the nature of the case; the gravity of the error; the ends of justice in the particular case; how the error came to the court’s attention; and whether the policies behind the general rule requiring preservation of error have been served in the case in another way, i.e., whether the trial court was, in some manner, presented with both sides of the issue and given an opportunity to correct any error.”

Id. at 382 n 6.

Sixteen years after the Supreme Court issued its opinion in Ailes, the court expressed its continued agreement with, and intent to “adhere to that statement” in Ailes, which the court characterized as setting out “precisely what [the Supreme Court] expects” when this court considers whether to exercise its discretion to correct plain error. State v. Fults, 343 Or 515, 521-22, 173 P3d 822 (2007). *929The Supreme Court emphasized that the requirement that this court articulate its reasons for exercising its discretion is “‘not a requirement of mere form.’” Id. at 522 (quoting Ailes, 312 Or at 382). In that context, the court observed, considerations that may factor into the discretionary analysis include “the interest of the judicial system in avoiding unnecessary repetitive sentencing proceedings, as well as its interest in requiring preservation of error.” Id. at 523.

More recently, the Supreme Court again emphasized that “a decision to review a plain error is one to be made with the ‘utmost caution’ because such review undercuts the policies served by the preservation doctrine.” Vanornum, 354 Or at 630-31. The court discussed those policies in detail in Peeples v. Lampert, 345 Or 209, 191 P3d 637 (2008):

“Those policies are prudential in nature. Preservation gives a trial court the chance to consider and rule on a contention, thereby possibly avoiding an error altogether or correcting one already made, which in turn may obviate the need for an appeal. Preservation also ensures fairness to an opposing party, by permitting the opposing party to respond to a contention and by otherwise not taking the opposing party by surprise. Finally, preservation fosters full development of the record, which aids the trial court in making a decision and the appellate court in reviewing it. Our jurisprudence, thus, has embraced the preservation requirement, not to promote form over substance but to promote an efficient administration of justice and the saving of judicial time.”

Id. at 219-20 (internal quotation marks, brackets, and citations omitted). Yet again, the court emphasized that, even when there is “an error apparent on the record, about which there is no reasonable dispute,” an appellate court “has discretion to consider [that] error, but it must do so with the ‘utmost caution,’ because of the strong policy reasons favoring preservation.” Id. at 219.

With those principles in mind, we first consider the gravity of any error that could be said to have plainly occurred. See Vanornum, 354 Or at 630 (citing “the gravity of the error” as one consideration in the discretionary analysis); Ailes, 312 Or at 382 n 6 (same). In determining whether *930any error was grave, we consider the likelihood that the error affected the outcome of the proceeding below. State v. Ramirez, 343 Or 505, 513, 173 P3d 817 (2007). Thus, we explained in State v. Pergande, 270 Or App 280, 348 P3d 245 (2015), that we would correct the trial court’s plain error in admitting testimony that commented on the credibility of two children who allegedly were abuse victims because that testimony likely influenced the verdict:

“As we have stated before, in a case that boils down to a credibility contest between the defendant and the victim, as here: evidence commenting on the credibility of either was likely to be harmful. [The witness’s] testimony was a direct comment on the credibility of [the children]. That error was grave because this is a sexual abuse case with no physical evidence of abuse and [the witness] was presented as an expert in treating sexual abuse victims with significant experience in spotting indications of suggestion or coaching. [The witness] was also the long-term counselor of both of the girls and familiar with them. Those facts present a significant risk that the jury’s credibility determinations were affected by [her] testimony.”

Id. at 285-86 (internal quotation marks and citations omitted); see also State v. Roelle, 259 Or App 44, 50, 312 P3d 555 (2013) (discussing the likely effect of erroneous admission of “prior bad acts” evidence in explaining that the error was grave).

Again, this is the testimony by McCourt that defendant contends the trial court should have stricken sua sponte:

“A. *** We like to talk to the witness again before the trial just to see, make sure they’re going to be able to — if they’re going to be able to come testify in a court, make sure they still remember, or if they have — if they decide— Let’s say that they were lying to us or they told us a fib, we like to find that out before we get to this point so we can — we don’t waste your time and the court’s time. And we talked to her on the Thursday of last week.

“Q. Okay. And if there had been any inconsistencies or any problems, wouldn’t those have needed to be reported to all parties?

“A. Yes.

*931“Q. And you did not write any such a report or * * * note any such inconsistencies at any time?

“A. I did not.”

(Emphasis added.)

The emphasized part of that testimony was objectionable, in that it could be understood to vouch for J’s credibility.2 That is, the statement could have conveyed to the jury that McCourt did not believe that J had lied or told a fib; otherwise, he would have taken steps to prevent the court’s time from being “waste [d].” See Corkill, 262 Or App at 552 (“true ‘vouching’ evidence” is a “witness’s testimony that he or she believes that another witness is or is not credible, which a party offers to bolster or undermine the veracity of that other witness”).

However, we reject defendant’s contention that, in addition, McCourt plainly vouched for J’s veracity when he gave testimony implying that there were not “any inconsistencies or any problems” in the child’s statements. “[Generally speaking, testimony that a witness’s statements were consistent with earlier statements that the witness made does not impermissibly vouch for the witness’s credibility.” State v. Beauvais, 357 Or 524, 547, 354 P3d 680 (2015); see also State v. Viranond, 346 Or 451, 460, 212 P3d 1252 (2009) (rejecting argument that one witness’s testimony that statements by another witness “were consistent with one another serves no real purpose other than indirectly to bolster” that other witness’s credibility; observing that, at least in some contexts, “testimony that witnesses’ earlier statements were consistent with their trial testimony merely established that the witnesses told the same story, true or false, more than once”). Thus, McCourt’s acknowledgement that he had not noticed any inconsistencies in J’s statements was not objectionable. True, we might have considered the prosecutor’s *932reference to other, unspecified “problems” to be problematic had defendant objected to testimony on that point and had the trial court overruled that objection. But given Beauvais and Viranond, we reject defendant’s argument that McCourt’s testimony that there were not “any inconsistencies or any problems” in J’s statements plainly was inadmissible and that the trial court plainly erred in not excluding it sua sponte.

However, that latter comment does provide context when we consider the gravity of any error that the trial court may have plainly committed when it did not strike McCourt’s testimony suggesting that he spoke with J to make sure that she had not fibbed. In that regard, we first observe that, although McCourt’s “fibbing” testimony could be understood to vouch for J’s credibility, he did not purport to base that opinion on any particular analysis of the child’s veracity. If anything, the prosecutor’s repeated references to “inconsistencies” in her next questions to McCourt suggest that McCourt’s opinion that J had not “fibbed” was based on little more than his belief that the child’s story had remained consistent over time.

That is not the kind of expert vouching testimony that most often has prompted this court to reverse a criminal conviction or a judgment favoring a civil plaintiff because the trial court should have stricken vouching testimony sua sponte. For example, we have reversed when trial courts failed to sua sponte strike testimony delivered by: a witness who was “presented as an expert in treating sexual abuse victims with significant experience in spotting indications of suggestion or coaching”;3 a mental-health counselor and a psychosexual evaluator who “emphatically and repeatedly offered their * * * opinions that [the] plaintiff was truthful in her accounts of the alleged abuse”;4 or a detective whose testimony “was couched in terms of his expertise in identifying untruthfulness.”5 Thus, we have explained, “In many cases *933where credibility is critical to the outcome, even a single ‘vouching’ statement by a witness * * * with years of experience and training in the field of child abuse prevention, can be given considerable weight by the jury.” State v. Ross, 271 Or App 1, 7, 349 P3d 620, rev den, 357 Or 743 (2015).6

This is not such a case. McCourt was not presented as an expert in identifying untruthfulness; nor was he presented as an expert in the field of childhood sexual abuse. In our view, his identity as a law-enforcement officer, standing alone, would not necessarily have led the jury to believe that he had special insight into J’s veracity. Moreover, as explained above, McCourt’s statements suggesting that J had not “fibbed” were closely associated with his agreement that he had not observed inconsistencies or unspecified “problems” in the child’s statements, suggesting that his opinion of the child’s veracity was based on little more than the unchanging nature of her story — a fact to which McCourt permissibly could testify. Accordingly, it is unlikely that McCourt’s “fibbing” testimony would have carried the same weight with the jury as the expert vouching testimony at issue in the cases discussed above. In addition, McCourt stated only once, and briefly, that he had interviewed J shortly before trial so he could make sure that she was not lying, and the prosecutor promptly redirected McCourt to permissible topics. Consequently, McCourt’s vouching testimony was not the sort of emphatic or repeated comment on credibility that has sometimes prompted this court to reverse because the trial court should have stricken the testimony sua sponte. In short, any error that occurred here is not nearly as grave as the errors that prompted reversals in those other cases.

We next consider “the nature of the case.” Ailes, 312 Or at 382 n 6. In that regard, defendant argues that J’s credibility was “at the heart of the state’s case” and “the totality of the evidence was hardly overwhelming.” We disagree. This is not a case, like some, in which the complainant’s testimony constitutes the sole direct evidence of sexual abuse. *934To the contrary, J’s report that defendant had licked her vagina was confirmed by Q, who testified that he had seen defendant bent down with his face near J’s private parts, asking for permission to lick her there. And, although E did not testify about that incident of abuse, she confirmed J’s report that defendant had looked under E’s dress. Thus, this case does not “boil[] down to a credibility contest between the defendant and the victim.” See Pergande, 270 Or App at 285 (describing a case that did present such a credibility contest). That fact, too, counsels against an exercise of Ailes discretion. See State v. Pickett, 246 Or App 62, 64-65, 264 P3d 209 (2011), rev den, 341 Or 541 (2012) (declining to correct plain error in admitting a diagnosis of sexual abuse in the absence of physical evidence of abuse because admission of the diagnosis “did not likely affect the court’s verdict” in the case, which “did not amount to merely a 'swearing match’ between the victim *** and defendant”; rather, the record included additional evidence of the defendant’s guilt, including his admission to having abused the victim).

We also consider “whether the policies behind the general rule requiring preservation of error have been served in the case in another way.” Ailes, 312 Or at 382 n 6. They have not. Had defendant objected to McCourt’s testimony, the trial court could have easily cured any error by striking the testimony and instructing the jury to disregard it. That consideration has not prevented us from exercising our discretion to correct plain “vouching error” in cases where the comment on credibility was so egregious that it was not clear that a curative instruction would have been sufficient. See State v. Higgins, 258 Or App, 177, 182, 308 P3d 352 (2013). Here, however, where McCourt’s vouching testimony was brief and was not delivered by an expert witness or other person with particular insight into the victim’s veracity, a curative instruction almost certainly would have sufficed had defendant objected to the testimony. See State v. Fulmer, 229 Or App 386, 395, 211 P3d 942 (2009), rev den, 348 Or 13 (2010) (“[I]f a trial court does erroneously allow a witness to comment on the credibility of another witness, reversal is not required if the trial court directs the jury to disregard the inappropriate testimony.”). In other words, *935any retrial that we might require in this case — with its attendant toll on the participants and the judicial system as a whole — would have been avoided entirely had defendant simply objected to McCourt’s testimony at the time it was delivered.

The judicial system’s interest in avoiding that kind of “unnecessary repetitive legal proceeding!], as well as its interest in requiring preservation of error,” is weighty. Dept. of Human Services v. E. L. G., 270 Or App 308, 315, 347 P3d 825 (2015). That principle is illustrated by the Supreme Court’s decision to affirm in State v. Cox, 337 Or 477, 98 P3d 1103 (2004), cert den, 546 US 830 (2005), an aggravated murder case in which the defendant was sentenced to death. In Cox, the defendant “identifie[d] 15 instances during his trial in which witnesses testified concerning another person’s out-of-court statements.” Id. at 500. The defendant argued “that admitting this testimony violate [d] his rights under the federal Confrontation Clause.” Id. He acknowledged that he had not objected to the evidence at trial, but argued that the Supreme Court should exercise its discretion to correct what he characterized as plain error. Id. The court declined to do so, even assuming that constitutional violations had occurred:

“Even if we assume that the error is plain, this is not an appropriate occasion to reach it. As the state notes, if defendant had raised a timely objection, the state could have found other ways to prove the facts that defendant now challenges, or it could have chosen to forgo the testimony and avoid the issue. In these circumstances, we decline to exercise our discretion to reach the unpreserved issues that defendant asks us to decide.”

Id. Here, too, the state easily could have decided not to ask McCourt any further questions about his pretrial conversation with J had defendant successfully objected to the “fibbing” testimony at trial. We understand from Cox, as well as cases like Vanornum, Fults, and Peeples, discussed above, that the ease with which any error could have been avoided or corrected should be a significant factor in an appellate court’s decision whether to exercise its discretion to correct a plain, but unpreserved, error.

*936And that brings us to the last factor that we deem pertinent to our decision in this case: “the ends of justice in the particular case.” Ailes, 312 Or at 382 n 6. Certainly, defendant has a compelling interest in not being wrongfully convicted of a serious crime. See State v. Reynolds, 250 Or App 516, 522, 280 P3d 1046, rev den, 352 Or 666 (2012). But that is true in virtually every case in which defendants raise unpreserved challenges to their criminal convictions. And in this case, we do not believe that the ends of justice weigh in favor of reversal. As explained, we have determined that the error in this case was not particularly grave, in terms of its likelihood of affecting the verdict, that the case did not amount to a simple swearing match between defendant and the victim, that any need for a retrial could have been easily avoided, and that the judicial system has a strong interest in avoiding such a retrial. No one of those factors, considered alone, necessarily would persuade us not to exercise our discretion to correct any plain error that might have occurred. When we consider those factors together, however, along with the Supreme Court’s admonition to exercise the “utmost caution” in addressing unpre-served claims of error, we conclude that the ends of justice militate against such an exercise of discretion in this case.

The dissent’s contrary conclusion makes several points to which we briefly respond. First, the dissent suggests that admission of McCourt’s vouching testimony could not be characterized as “harmless” in the context of this case, given the lack of physical evidence of abuse and the inconsistencies in the children’s descriptions of events. See 275 Or App at 957-58 (Sercombe, J., dissenting). That may be. That is, any error in admitting McCourt’s testimony might not be deemed harmless if we were considering its effect in the context of a preserved evidentiary objection. But the harmless-error analysis does not govern our discretionary decision about whether to address unpreserved claims of error. Instead, we must balance the gravity of any error, in the context of the “nature of the case,” against the other factors set forth in Ailes, Vanornum, Fults, and other plain-error cases. This decision reflects our view of the appropriate weighing of those factors.

*937The dissent also downplays both the importance of preservation principles — a point which, as we have noted, the Supreme Court continues to emphasize — and the judicial system’s interest in avoiding repetitive proceedings. In doing so, the dissent refers to Reynolds, 250 Or App at 524, a case in which we exercised our discretion to address the defendant’s unpreserved claim that his conviction for third-degree assault was not based on constitutionally sufficient evidence. 275 Or App at 954-55 (Sercombe, J., dissenting). The comparison to Reynolds is inapt. In that case, the state conceded that the evidence in the record did not support the challenged conviction. 250 Or App at 518. Moreover, we concluded that, had the defendant moved for a judgment of acquittal in the trial court, it was “unlikely that the record would have developed differently,” noting that the state did “not contend otherwise.” Id. at 524. Under those circumstances, it was, perhaps, unsurprising that we rejected “the ‘interest in requiring preservation of error’ factor” as dispos-itive in Reynolds, as the dissent states here. 275 Or App at 954-55 (Sercombe, J., dissenting).

This case differs from Reynolds in two significant respects. First, the result in Reynolds was straightforward: The defendant’s conviction for third-degree assault was reversed and remanded for entry of a judgment of conviction for, instead, fourth-degree assault. 250 Or App at 527. In this case, reversal would be followed by a new trial in which, at least theoretically, the evidence presented over two days of trial — including the testimony delivered by the child witnesses — might be essentially the same as it was at the original trial held in January 2013, except that McCourt would not testify that he had spoken with J before trial to ensure that she had not lied or told a “fib.” The judicial system’s — and the children’s — interest in avoiding that kind of retrial cannot be equated to the system’s interest in avoiding a simple remand for entry of a less-serious conviction, as in Reynolds.

Second, as noted above, we emphasized in Reynolds that the evidentiary record would not have developed differently had the defendant preserved his claim of error; that is, the state apparently did not contend that it might have been able to present evidence supporting a conviction for *938third-degree assault. Here, though, the record presumably would have developed differently had defendant objected to McCourt’s testimony. The trial court presumably would have stricken that testimony and instructed the jury to disregard it.

The dissent concludes that that presumptive difference in how the record would have developed, had defendant objected to McCourt’s testimony on “vouching” grounds, is so significant that defendant “would likely” prevail if he pursued post-conviction relief on the theory that his trial lawyer should have made that objection. 275 Or App at 960-61 (Sercombe, J., dissenting). But the ends of justice do not require a retrial simply because some view it as “likely” that defendant would prevail in a later post-conviction case. The necessarily limited scope of the record on direct appeal makes it difficult, if not impossible, for us to predict the eventual result in any such collateral proceeding. On this record, we know nothing about why defendant’s lawyer did not object to McCourt’s testimony. Nor can we adequately assess whether — perhaps for reasons that are not reflected in the trial-court record — a decision not to object could have been a choice within the wide range of options available to a reasonably competent lawyer under the circumstances then presented, or whether defendant was prejudiced by that kind of decision. If defendant’s lawyer acted reasonably by not objecting to McCourt’s testimony, or if defendant was not prejudiced by the absence of an objection, then justice would not demand a retrial based on the trial court not having stricken that testimony sua sponte.

In the end, we cannot conclude from the record on direct appeal that there is no meaningful chance either that defendant’s lawyer exercised reasonable professional skill and judgment when he did not object to McCourt’s testimony or that, given the totality of the circumstances as reflected in a yet-to-be-developed post-conviction record, defendant was not prejudiced by the absence of an objection. The purpose of post-conviction litigation is to allow development of a record that permits a factfinder to make fully informed decisions on those points. See Pereida-Alba v. Coursey, 356 Or 654, 661-62, 342 P3d 70 (2015) (describing elements of a post-conviction claim based on an inadequate-assistance *939theory). Defendant’s remedy — if any — for the lack of an objection to McCourt’s testimony must be in that type of collateral proceeding.

Affirmed.

Haselton, C. J., and Ortega, DeVore, and Garrett, JJ., join in this opinion.